Mapping and Sequencing the Human Genome
Mapping and Sequencing the Human Genome
A primary goal of the Human Genome Project is to make a series of descriptive diagrams mapsof each human chromosome at increasingly finer resolutions. Mapping involves (1) dividing the chromosomes into smaller fragments that can be propagated and characterized and (2) ordering (mapping) them to correspond to their respective locations on the chromosomes. After mapping is completed, the next step is to determine the base sequence of each of the ordered DNA fragments. The ultimate goal of genome research is to find all the genes in the DNA sequence and to develop tools for using this information in the study of human biology and medicine. Improving the instrumentation and techniques required for mapping and sequencinga major focus of the genome projectwill increase efficiency and cost-effectiveness. Goals include automating methods and optimizing techniques to extract the maximum useful information from maps and sequences.
A genome map describes the order of genes or other markers and the spacing between them on each chromosome. Human genome maps are constructed on several different scales or levels of resolution. At the coarsest resolution are genetic linkage maps, which depict the relative chromosomal locations of DNA markers (genes and other identifiable DNA sequences) by their patterns of inheritance. Physical maps describe the chemical characteristics of the DNA molecule itself.
Geneticists have already charted the approximate positions of over 2300 genes, and a start has been made in establishing high-resolution maps of the genome (Fig. 7). More-precise maps are needed to organize systematic sequencing efforts and plan new research directions.
HUMAN GENOME PROJECT GOALS Resolution
- Complete a detailed human genetic map
- Complete a physical map
- Acquire the genome as clones
- Determine the complete sequence
- Find all the genes
With the data generated by the project, investigators will determine the functions of the genes and develop tools for biological and medical applications.
Genetic Linkage Maps
A genetic linkage map shows the relative locations of specific DNA markers along the chromosome. Any inherited physical or molecular characteristic that differs among individuals and is easily detectable in the laboratory is a potential genetic marker. Markers can be expressed DNA regions (genes) or DNA segments that have no known coding function but whose inheritance pattern can be followed. DNA sequence differences are especially useful markers because they are plentiful and easy to characterize precise ly.
Markers must be polymorphic to be useful in mapping; that is, alternative forms must exist among individuals so that they are detectable among different members in family studies. Polymorphisms are variations in DNA sequence that occur on average once every 300 to 500 bp. Variations within exon sequences can lead to observable changes, such as differences in eye color, blood type, and disease susceptibility. Most variations occur within introns and have little or no effect on an organisms appearance or function, yet they are detectable at the DNA level and can be used as markers. Examples of these types of markers include (1)_restriction fragment length polymorphisms (RFLPs), which reflect sequence variations in DNA sites that can be cleaved by DNA restriction enzymes (see box, p. 203), and (2)_variable number of tandem repeat sequences, which are short repeated sequences that vary in the number of repeated units and, therefore, in length (a characteristic easily measured). The human genetic linkage map i s constructed by observing how frequently two markers are inherited together.
Two markers located near each other on the same chromosome will tend to be passed together from parent to child. During the normal production of sperm and egg cells, DNA strands occasionally break and rejoin in different places on the same chromosome or on the other copy of the same chromosome (i.e., the homologous chromosome). This process (called meiotic recombination) can result in the separation of two markers originally on the same chromosome (Fig. 8). The closer the markers are to each other the more tightly linkedthe less likely a recombination event will fall between and separate them. Recombination frequency thus provides an estimate of the distance between two markers.
On the genetic map, distances between markers are measured in terms of centimorgans (cM), named after the American geneticist Thomas Hunt Morgan. Two markers are said to be 1_cM apart if they are separated by recombination 1% of the time. A genetic distance of 1_cM is roughly equal to a physical distance of 1 million bp (1 Mb). The current resolution of most human genetic map regions is about 10 Mb.
The value of the genetic map is that an inherited disease can be located on the map by following the inheritance of a DNA marker present in affected individuals (but absent in unaffected individuals), even though the molecular basis of the disease may not yet be understood nor the responsible gene identified. Genetic maps have been used to find the exact chromosomal location of several important disease genes, including cystic fibrosis, sickle cell disease, Tay-Sachs disease, fragile X syndrome, and myotonic dystrophy.
One short-term goal of the genome project is to develop a high- resolution genetic map (2 to 5_cM); recent consensus maps of some chromosomes have averaged 7 to 10_cM between genetic markers. Genetic mapping resolution has been increased through the application of recombinant DNA technology, including in vitro radiati on- induced chromosome fragmentation and cell fusions (joining human cells with those of other species to form hybrid cells) to create panels of cells with specific and varied human chromosomal components. Assessing the frequency of marker sites remaining together after radiation-induced DNA fragmentation can establish the order and distance between the markers. Because only a single copy of a chromosome is required for analysis, even nonpolymorphic markers are useful in radiation hybrid mapping. [In meiotic mapping (described above), two copies of a chromosome must be distinguished from each other by polymorphic markers.]
Restriction Enzymes: Microscopic Scalpels
Isolated from various bacteria, restriction enzymes recognize short DNA sequences and cut the DNA molecules at those specific sites. (A natural biological function of these enzymes is to protect bacteria by attacking viral and other foreign DNA.) Some res triction enzymes (rare-cutters) cut the DNA very infrequently, generating a small number of very large fragments (several thousand to a million bp). Most enzymes cut DNA more frequently, thus generating a large number of small fragments (less than a hundred to more than a thous and bp).
On average, restriction enzymes with
* 4-base recognition sites will yield pieces 256 bases long,
* 6-base recognition sites will yield pieces 4000 bases long, and
* 8-base recognition sites will yield pieces 64,000 bases long.
Since hundreds of different restriction enzymes have been characterized, DNA can be cut into many different small fragments.
Different types of physical maps vary in their degree of resolution. The lowest-resolution physical map is the chromosomal (sometimes called cytogenetic) map, which is based on the distinctive banding patterns observed by light microscopy of stained chromosomes. A cDNA map shows the locations of expressed DNA regions (exons) on the chromosomal map. The more detailed cosmid contig map depicts the order of overlapping DNA fragments spanning the genome. A macrorestriction map describes the order and distance between enzyme cutting (cleavage) sites. The highest- resolution physical map is the complete elucidation of the DNA base-pair sequence of each chromosome in the human genome. Physical maps are described in greater detail below.
Low-Resolution Physical Mapping
Chromosomal map. In a chromosomal map, genes or other identifiable DNA fragments are assigned to their respective chromosomes, with distances measured in base pairs. These markers can be physically associated with particular bands (identified by cytogenetic staining) primarily by in situ hybridization, a technique that involves tagging the DNA marker with an observable label (e.g., one that fluoresces or is radioactive). The location of the labeled probe can be detected after it binds to its comple mentary DNA strand in an intact chromosome.
As with genetic linkage mapping, chromosomal mapping can be used to locate genetic markers defined by traits observable only in whole organisms. Because chromosomal maps are based on estimates of physical distance, they are considered to be physical maps. The number of base pairs within a band can only be estimated.
Until recently, even the best chromosomal maps could be used to locate a DNA fragment only to a region of about 10 Mb, the size of a typical band seen on a chromosome. Improvements in fluorescence in situ hybridization (FISH) methods allow orientation of DNA sequences that lie as close as 2 to 5 Mb. Modifications to in situhybridization methods, using chromosomes at a stage in cell division (interphase) when they are less compact, increase map resolution to around 100,000 bp. Further banding refinement might allow chromosomal bands to be associated with specific amplified DNA fragments, an improvement that could be useful in analyzing observable physical traits associated with chromosomal abnormalities.
cDNA map. A cDNA map shows the positions of expressed DNA regions (exons) relative to particular chromosomal regions or bands. (Expressed DNA regions are those transcribed into mRNA.) cDNA is synthesized in the laboratory using the mRNA molecule as a template; base-pairing rules are followed (i.e., an A on the mRNA molecule will pair with a T on the new DNA strand). This cDNA can then be mapped to genomic regions.
Because they represent expressed genomic regions, cDNAs are thought to identify the parts of the genome with the most biological and medical significance. A cDNA map can provide the chromosomal location for genes whose functions are currently unknown. For disease-gene hunters, the map can also suggest a set of candidate genes to test when the approximate location of a disease gene has been mapped by genetic linkage techniques.
High-Resolution Physical Mapping
The two current approaches to high-resolution physical mapping are termed top- down (producing a macrorestriction map) and bottom-up (resulting in a contig map). With either strategy (described below) the maps represent ordered sets of DNA fragments that are generated by cutting genomic DNA with restriction enzymes (see previously discussed Restriction Enzymes). The fragments are then amplified by cloning or by polymerase chain reaction (PCR) methods (see DNA Amplification below). Electrophoretic techniques are used to separate the fragments according to size into different bands, which can be visualized by direct DNA staining or by hybridization with DNA probes of interest. The use of purified chromosomes separated either by flow sorting from human cell lines or in hybrid cell lines allows a single chromosome to be mapped (see Separating Chromosomes below).
A number of strategies can be used to reconstruct the original order of the DNA fragments in the genome. Many approaches make use of the ability of single strands of DNA and/or RNA to hybridizeto form double-stranded segments by hydrogen bonding between complementary bases. The extent of sequence homology between the two strands can be inferred from the length of the double-stranded segment. Fingerprinting uses restriction map data to determine which fragments have a specific sequence (fingerprint) in common and therefore overlap. Another approach uses linking clones as probes for hybridization to chromosomal DNA cut with the same restriction enzyme.
Macrorestriction maps: Top-down mapping. In top-down mapping, a single chromosome is cut (with rare-cutter restriction enzymes) into large pieces, which are ordered and subdivided; the smaller pieces are then mapped further. The resulting macro-restriction maps depict the order of and distance between sites at which rare-cutter enzymes cleave (Fig. 9a). This approach yields maps with more continuity and fewer gaps between fragments than contig maps (see below), but map resolution is lower and may not be useful in finding particular genes; in addition, this strategy generally does not produce long stretches of mapped sites. Currently, this approach allows DNA pieces to be located in regions measuring about 100,000 bp to 1_Mb.
The development of pulsed-field gel (PFG) electrophoretic methods has improved the mapping and cloning of large DNA molecules. While conventional gel electrophoretic methods separate pieces less than 40 kb (1 kb = 1000 bases) in size, PFG separates molecules up to 10 Mb, allowing the application of both conventional and new mapping methods to larger genomic regions.
Contig maps: Bottom-up mapping. The bottom-up approach involves cutting the chromosome into small pieces, each of which is cloned and ordered. The ordered fragments form contiguous DNA blocks (contigs). Currently, the resulting library of clones varies in size from 10,000 bp to 1 Mb (Fig. 9b). An advantage of this approach is the accessibility of these stable clones to other researchers. Contig construction can be verified by FISH, which localizes cosmids to specific regions within chromosomal bands.
Contig maps thus consist of a linked library of small overlapping clones representing a complete chromosomal segment. While useful for finding genes localized to a small area (under 2 Mb), contig maps are difficult to extend over large stretches of a chromosome because all regions are not clonable. DNA probe techniques can be used to fill in the gaps, but they are time consuming. Figure 10 is a diagram relating the different types of maps.
Technological improvements now make possible the cloning of large DNA pieces, using artificially constructed chromosome vectors that carry human DNA fragments as large as 1 Mb. These vectors are maintained in yeast cells as artificial chromosomes (YACs). For more explanation, see DNA Amplification below) Before YACs were developed, the largest cloning vectors (cosmids) carried inserts of only 20 to 40 kb. YAC methodology drastically reduces the number of clones to be ordered; many YACs span entire human genes. A more detailed map of a large YAC insert can be produced by subcloning, a process in which fragments of the original insert are cloned into smaller-insert vectors. Because some YAC regions are unstable, large-capacity bacterial vectors (i.e., those that can accommodate large inserts) are also being developed.
Pioneered at Los Alamos National Laboratory (LANL), flow sorting employs flow cytometry to separate, according to size, chromosomes isolated from cells during cell division when they are condensed and stable. As the chromosomes flow singly past a laser beam, they are differentiated by analyzing the amount of DNA present, and individual chromosomes are directed to specific collection tubes.
Somatic cell hybridization
In somatic cell hybridization, human cells and rodent tumor cells are fused (hybridized); over time, after the chromosomes mix, human chromosomes are preferentially lost from the hybrid cell until only one or a few remain. Those individual hybrid cells are then propagated and maintained as cell lines containing specific human chromosomes. Improvements to this technique have generated a number of hybrid cell lines, each with a specific single human chromosome.
The ultimate physical map of the human genome is the complete DNA sequencethe determination of all base pairs on each chromosome. The completed map will provide biologists with a Rosetta stone for studying human biology and enable medical researchers to begin to unravel the mechanisms of inherited diseases. Much effort continues to be spent locating genes; if the full sequence were known, emphasis could shift to determining gene function. The Human Genome Project is creating research tools for 21st-century biology, when the goal will be to understand the sequence and functions of the genes residing therein.
Achieving the goals of the Human Genome Project will require substantial improvements in the rate, efficiency, and reliability of standard sequencing procedures. While technological advances are leading to the automation of standard DNA purification, separation, and detection steps, efforts are also focusing on the development of entirely new sequencing methods that may eliminate some of these steps. Sequencing procedures currently involve first subcloning DNA fragments from a cosmid or bacteriophage library into special sequencing vectors that carry shorter pieces of the original cosmid fragments (Fig. 11). The next step is to make the subcloned fragments into sets of nested fragments differing in length by one nucleotide, so that the specific base at the end of each successive fragment is detectable after the fragments have been separated by gel electrophoresis. Current sequencing technologies are discussed later.
DNA Amplification: Cloning and Polymerase Chain Reaction
Cloning (in vivo DNA amplification)
Cloning involves the use of recombinant DNA technology to propagate DNA fragments inside a foreign host. The fragments are usually isolated from chromosomes using restriction enzymes and then united with a carrier (a vector). Following introduction into suitable host cells, the DNA fragments can then be reproduced along with the host cell DNA. Vectors are DNA molecules originating from viruses, bacteria, and yeast cells. They accommodate various sizes of foreign DNA fragments ranging from 12,000 bp for bacterial vectors (plasmids and cosmids) to 1 Mb for yeast vectors (yeast artificial chromosomes). Bacteria are most often the hosts for these inserts, but yeast and mammalian cells are also used. (Figure 11a: Cloning DNA in Plasmids)
Cloning procedures provide unlimited material for experimental study. A random (unordered) set of cloned DNA fragments is called a library. Genomic libraries are sets of overlapping fragments encompassing an entire genome Figure 11b: Cloning DNA in Plasmids. Also available are chromosome-specific libraries, which consist of fragments derived from source DNA enriched for a particular chromosome. (See Separating Chromosomes, above.)
PCR (in vitro DNA amplification)
Described as being to genes what Gutenberg's printing press was to the written word, PCR can amplify a desired DNA sequence of any origin (virus, bacteria, plant, or human) hundreds of millions of times in a matter of hours, a task that would have required several days with recombinant technology. PCR is especially valuable because the reaction is highly specific, easily automated, and capable of amplifying minute amounts of sample. For these reasons, PCR has also had a major impact on clinical medicine, genetic disease diagnostics, forensic science, and evolutionary biology.
PCR is a process based on a specialized polymerase enzyme, which can synthesize a complementary strand to a given DNA strand in a mixture containing the 4 DNA bases and 2 DNA fragments (primers, each about 20 bases long) flanking the target sequence. The mixture is heated to separate the strands of double-stranded DNA containing the target sequence and then cooled to allow (1) the primers to find and bind to their complementary sequences on the separated strands and (2) the polymerase to extend the primers into new complementary strands. Repeated heating and cooling cycles multiply the target DNA exponentially, since each new double strand separates to become two templates for further synthesis. In about 1 hour, 20 PCR cycles can amplify the target by a millionfold.
Current Sequencing Technologies
The two basic sequencing approaches, Maxam-Gilbert and Sanger, differ primarily in the way the nested DNA fragments are produced. Both methods work because gel electrophoresis produces very high resolution separations of DNA molecules; even fragments that differ in size by only a single nucleotide can be resolved. Almost all steps in these sequencing methods are now automated. Maxam-Gilbert sequencing (also called the chemical degradation method) uses chemicals to cleave DNA at specific bases, resulting in fragments of different lengths. A refinement to the Maxam-Gilbert method known as multiplex sequencing enables investigators to analyze about 40 clones on a single DNA sequencing gel. Sanger sequencing (also called the chain termination or dideoxy method) involves using an enzymatic procedure to synthesize DNA chains of varying length in four different reactions, stopping the DNA replication at positions occupied by one of the four bases, and then determining the resulting fragment lengths (Fig. 12).
These first-generation gel-based sequencing technologies are now being used to sequence small regions of interest in the human genome. Although investigators could use existing technology to sequence whole chromosomes, time and cost considerations make large- scale sequencing projects of this nature impractical. The smallest human chromosome (Y) contains 50 Mb; the largest (chromosome 1) has 250 Mb. The largest continuous DNA sequence obtained thus far, however, is approximately 350,000 bp, and the best available equipment can sequence only 50,000 to 100,000 bases per year at an approximate cost of $1 to $2 per base. At that rate, an unacceptable 30,000 work- years and at least $3_billion would be required for sequencing alone.
Sequencing Technologies Under Development
A major focus of the Human Genome Project is the development of automated sequencing technology that can accurately sequence 100,000 or more bases per day at a cost of less than $.50 per base. Specific goals include the development of sequencing and detection schemes that are faster and more sensitive, accurate, and economical. Many novel sequencing technologies are now being explored, and the most promising ones will eventually be optimized for widespread use.
Second-generation (interim) sequencing technologies will enable speed and accuracy to increase by an order of magnitude (i.e., 10 times greater) while lowering the cost per base. Some important disease genes will be sequenced with such technologies as (1) high-voltage capillary and ultrathin electrophoresis to increase fragment separation rate and (2) use of resonance ionization spectroscopy to detect stable isotope labels.
Third-generation gel-less sequencing technologies, which aim to increase efficiency by several orders of magnitude, are expected to be used for sequencing most of the human genome. These developing technologies include (1) enhanced fluorescence detection of individual labeled bases in flow cytometry, (2) direct reading of the base sequence on a DNA strand with the use of scanning tunneling or atomic force microscopies, (3) enhanced mass spectrometric analysis of DNA sequence, and (4) sequencing by hybridization to short panels of nucleotides of known sequence. Pilot large-scale sequencing projects will provide opportunities to improve current technologies and will reveal challenges investigators may encounter in larger- scale efforts.
Partial Sequencing To Facilitate Mapping, Gene Identification
Correlating mapping data from different laboratories has been a problem because of differences in generating, isolating, and mapping DNA fragments. A common reference system designed to meet these challenges uses partially sequenced unique regions (200 to 500 bp) to identify clones, contigs, and long stretches of sequence. Called sequence tagged sites (STSs), these short sequences have become standard markers for physical mapping.
Because coding sequences of genes represent most of the potentially useful information content of the genome (but are only a fraction of the total DNA), some investigators have begun partial sequencing of cDNAs instead of random genomic DNA. (cDNAs are de rived from mRNA sequences, which are the transcription products of expressed genes.) In addition to providing unique markers, these partial sequences [termed expressed sequence tags (ESTs)] also identify expressed genes. This strategy can thus provide a m eans of rapidly identifying most human genes. Other applications of the EST approach include determining locations of genes along chromosomes and identifying coding regions in genomic sequences.
End Games: Completing Maps and Sequences; Finding Specific Genes
Starting maps and sequences is relatively simple; finishing them will require new strategies or a combination of existing methods. After a sequence is determined using the methods described above, the task remains to fill in the many large gaps left by current mapping methods. One approach is single-chromosome microdissection, in which a piece is physically cut from a chromosomal region of particular interest, broken up into smaller pieces, and amplified by PCR or cloning (see DNA Amplification above). These fragments can then be mapped and sequenced by the methods previously described.
Chromosome walking, one strategy for filling in gaps, involves hybridizing a primer of known sequence to a clone from an unordered genomic library and synthesizing a short complementary strand (called walking along a chromosome). The complementary strand is then sequenced and its end used as the next primer for further walking; in this way the adjacent, previously unknown, region is identified and sequenced. The chromosome is thus systematically sequenced from one end to the other. Because primers must be synthesized chemically, a disadvantage of this technique is the large number of different primers needed to walk a long distance. Chromosome walking is also used to locate specific genes by sequencing the chromosomal segments between markers that flank t he gene of interest (Fig. 13).
The current human genetic map has about 1000 markers, or 1 marker spaced every 3_million bp; an estimated 100 genes lie between each pair of markers. Higher-resolution genetic maps have been made in regions of particular interest. New genes can be located by combining genetic and physical map information for a region. The genetic map basically describes gene order. Rough information about gene location is som etimes available also, but these data must be used with caution because recombination is not equally likely at all places on the chromosome. Thus the genetic map, compared to the physical map, stretches in some places and compresses in others, as though it were drawn on a rubber band.
The degree of difficulty in finding a disease gene of interest depends largely on what information is already known about the gene and, especially, on what kind of DNA alterations cause the disease. Spotting the disease gene is very difficult when disease results from a single altered DNA base; sickle cell anemia is an example of such a case, as are probably most major human inherited diseases. When disease results from a large DNA rearrangement, this anomaly can usually be detected as alterations in the physical map of the region or even by direct microscopic examination of the chromosome. The location of these alterations pinpoints the site of the gene.
Identifying the gene responsible for a specific disease without a map is analogous to finding a needle in a haystack. Actually, finding the gene is even more difficult, because even close up, the gene still looks like just another piece of hay. However, maps give clues on where to look; the finer the maps resolution, the fewer pieces of hay to be tested.
Once the neighborhood of a gene of interest has been identified, several strategies can be used to find the gene itself. An ordered library of the gene neighborhood can be constructed if one is not already available. This library provides DNA fragments that can be screened for additional polymorphisms, improving the genetic map of the region and further restricting the possible gene location. In addition, DNA fragments from the region can be used as probes to search for DNA sequences that are expressed (transcribed to RNA) or conserved among individuals. Most genes will have such sequences. Then individual gene candidates must be examined. For example, a gene responsible for liver disease is likely to be expressed in the liver and less likely in other tis sues or organs. This type of evidence can further limit the search. Finally, a suspected gene may need to be sequenced in both healthy and affected individuals. A consistent pattern of DNA variation when these two samples are compared will show that the gene of interest has very likely been found. The ultimate proof is to correct the suspected DNA alteration in a cell and show that the cells behavior reverts to normal.
Next Section: Model Organism Research
Go to the Table of Contents